Multiple autoimmune diseases including lupus (SLE), scleroderma (SSc), Sjogren's Syndrome (SS), rheumatoid arthritis (RA), autoimmune hepatitis (AH), and primary biliary cirrhosis (PBC) sometimes occur in the same person. The reason is unknown. All these diseases require multiple genetic susceptibility loci and an environmental contribution. Agents causing oxidative stress also associate with these diseases, but how this contributes to disease pathogenesis is unclear. T cells responding to host MHC molecules in chronic graft-vs-host disease (cGVHD) cause a similar disease spectrum. Why T cells responding to host MHC cause these diseases in cGVHD is unclear, but may be due to the host's genetic makeup. Inhibiting DNA methylation in CD4'' T cells makes them responsive to self MHC molecules, similar to the response causing cGVHD, and mice receiving the modified cells develop anti-dsDNA antibodies, with or without autoimmune disease depending on the strain. Oxidative stress also causes DNA demethylation, and T cells treated with oxidizers cause autoimmunity in mice. We have now found that the genes overexpressed by experimentally demethylated CD4+ T cells are all co-expressed on a novel CD4+ CD28+ KIR+CD70 hiCD11a hi CD40L hi subset. The same subset is found in patients with active SLE, SSc, SS, and RA. We hypothesize that this MHCresponsive subset contributes to human autoimmune diseases that resemble cGVHD. Since oxidative stress induces this subset in vitro, we also hypothesize that oxidative stress causes its development in patients with these autoimmune diseases. We will test these hypotheses by: 1. Comparing the size of the demethylated CD4+ D28+ KIR+ subset in patients with active and inactive systemic autoimmune diseases (SLE, SS, SSc, RA and autoimmune liver disease) to age, sex and ethnicity-matched controls and patients with active infections, and 2. Determining the relationship of subset size to disease activity and biomarkers of oxidative stress in these subjects and controls. Future studies would determine the relationship of disease manifestations to genetic predisposition.